Sources: DeepSeek and Moonshot AI heavily relied on GPT-6 Astra distillation to train models like Kimi K3.1 Pro and V4.1-Flash 1017, after Anthropic's crackdown (The Information)

AI 摘要

DeepSeek and Moonshot AI reportedly relied heavily on GPT-6 Astra distillation to train models such as Kimi K3.1 Pro and V4.1-Flash 1017. This reliance occurred after Anthropic's crackdown, with OpenAI's GPT-6 Astra, released in early September, experiencing strong demand as the company's most powerful model.